This paper presents a laser-based tracking of people using two sensor nodes fixed in an environments (infra-sensor nodes). Our system consists of two sensor nodes equipped with four-layered laser scanner (LS) and a central server. After the network among the server and sensor nodes is connected automatically, the laser measurements captured by the sensor nodes are collected to the central server. Based on the laser measurements, the relative posture of two sensor nodes is calculated based on Corresponding Vector Fitting Sample and Consensus (CVFSAC) method, and the server begins people detection and tracking; the server extracts the position data of people by background subtraction method. By using the position data of people, heuristic-rule-based and global-nearest-neighbor-based data association identifies multiple people in crowded environments. Their identified people are tracked via model-based tracker; the interacting-multi-model (IMM) estimator is applied to tracking people with sudden changes in motion; walking, running or stopping. Experimental results of tracking 18 people validate the effectiveness of our method.